Implement API key usage

.env.example

@@ -4,6 +4,7 @@
 # Server connection settings
 ORPHEUS_API_URL=http://127.0.0.1:1234/v1/completions
 ORPHEUS_API_TIMEOUT=120 # You should scale this value based on max tokens and your inference speed
+ORPHEUS_API_KEY=your_api_key_here # API key for authentication with the OpenAI-compatible API
 
 # Generation parameters
 ORPHEUS_MAX_TOKENS=8192 # If you want longer completions, increase this value

README.md

@@ -140,6 +140,7 @@ The server provides an OpenAI-compatible API endpoint at `/v1/audio/speech`:
 ```bash
 curl http://localhost:5005/v1/audio/speech \
   -H "Content-Type: application/json" \
+  -H "Authorization: Bearer your_api_key_here" \
   -d '{
     "model": "orpheus",
     "input": "Hello world! This is a test of the Orpheus TTS system.",
@@ -165,6 +166,7 @@ Additionally, a simpler `/speak` endpoint is available:
 ```bash
 curl -X POST http://localhost:5005/speak \
   -H "Content-Type: application/json" \
+  -H "Authorization: Bearer your_api_key_here" \
   -d '{
     "text": "Hello world! This is a test.",
     "voice": "tara"
@@ -265,7 +267,7 @@ You can easily integrate this TTS solution with [OpenWebUI](https://github.com/o
 2. In OpenWebUI, go to Admin Panel > Settings > Audio
 3. Change TTS from Web API to OpenAI
 4. Set APIBASE URL to your server address (e.g., `http://localhost:5005/v1`)
-5. API Key can be set to "not-needed"
+5. Set API Key to your configured API key or "not-needed" if API key authentication is not enabled
 6. Set TTS Voice to one of the available voices: `tara`, `leah`, `jess`, `leo`, `dan`, `mia`, `zac`, or `zoe`
 7. Set TTS Model to `tts-1`
 
@@ -295,6 +297,7 @@ You can configure the system using environment variables or a `.env` file:
 
 - `ORPHEUS_API_URL`: URL of the LLM inference API (tts_engine/inference.py)
 - `ORPHEUS_API_TIMEOUT`: Timeout in seconds for API requests (default: 120)
+- `ORPHEUS_API_KEY`: API key for authentication with the OpenAI-compatible API (optional)
 - `ORPHEUS_MAX_TOKENS`: Maximum tokens to generate (default: 8192)
 - `ORPHEUS_TEMPERATURE`: Temperature for generation (default: 0.6)
 - `ORPHEUS_TOP_P`: Top-p sampling parameter (default: 0.9)

app.py

@@ -7,6 +7,7 @@ import time
 import asyncio
 from datetime import datetime
 from typing import List, Optional
+from fastapi.security import HTTPBearer, HTTPAuthorizationCredentials
 from dotenv import load_dotenv
 
 # Function to ensure .env file exists
@@ -28,7 +29,7 @@ ensure_env_file_exists()
 # Load environment variables from .env file
 load_dotenv(override=True)
 
-from fastapi import FastAPI, Request, Form, HTTPException, Depends
+from fastapi import FastAPI, Request, Form, HTTPException, Depends, Security
 from fastapi.responses import HTMLResponse, FileResponse, JSONResponse
 from fastapi.staticfiles import StaticFiles
 from fastapi.templating import Jinja2Templates
@@ -59,6 +60,33 @@ app.mount("/static", StaticFiles(directory="static"), name="static")
 # Setup templates
 templates = Jinja2Templates(directory="templates")
 
+# API key authentication
+security = HTTPBearer(auto_error=False)
+
+# Get API key from environment
+API_KEY = os.environ.get("ORPHEUS_API_KEY")
+
+# Function to verify API key
+async def verify_api_key(credentials: HTTPAuthorizationCredentials = Security(security)):
+    """
+    Verify the API key from the Authorization header.
+    If no API key is configured, authentication is skipped.
+    """
+    # If no API key is configured, skip authentication
+    if not API_KEY:
+        return True
+    
+    # If API key is configured but no credentials provided, raise 401
+    if not credentials:
+        raise HTTPException(
+            status_code=401,
+            detail="Missing API key. Please provide an API key in the Authorization header."
+        )
+    
+    # Verify the API key
+    if credentials.credentials != API_KEY:
+        raise HTTPException(status_code=401, detail="Invalid API key")
+
 # API models
 class SpeechRequest(BaseModel):
     input: str
@@ -75,7 +103,7 @@ class APIResponse(BaseModel):
 
 # OpenAI-compatible API endpoint
 @app.post("/v1/audio/speech")
-async def create_speech_api(request: SpeechRequest):
+async def create_speech_api(request: SpeechRequest, authorized: bool = Depends(verify_api_key)):
     """
     Generate speech from text using the Orpheus TTS model.
     Compatible with OpenAI's /v1/audio/speech endpoint.
@@ -116,7 +144,7 @@ async def create_speech_api(request: SpeechRequest):
 
 # Legacy API endpoint for compatibility
 @app.post("/speak")
-async def speak(request: Request):
+async def speak(request: Request, authorized: bool = Depends(verify_api_key)):
     """Legacy endpoint for compatibility with existing clients"""
     data = await request.json()
     text = data.get("text", "")

tts_engine/inference.py

@@ -85,9 +85,17 @@ API_URL = os.environ.get("ORPHEUS_API_URL")
 if not API_URL:
     print("WARNING: ORPHEUS_API_URL not set. API calls will fail until configured.")
 
+# Get API key from environment
+API_KEY = os.environ.get("ORPHEUS_API_KEY")
+if not API_KEY:
+    print("WARNING: ORPHEUS_API_KEY not set. API calls may fail if the endpoint requires authentication.")
+
+# Set up headers with authentication if API key is provided
 HEADERS = {
     "Content-Type": "application/json"
 }
+if API_KEY:
+    HEADERS["Authorization"] = f"Bearer {API_KEY}"
 
 # Request timeout settings
 try:
@@ -252,6 +260,12 @@ def generate_tokens_from_api(prompt: str, voice: str = DEFAULT_VOICE, temperatur
             
             if response.status_code != 200:
                 print(f"Error: API request failed with status code {response.status_code}")
+                if response.status_code == 401:
+                    print(f"Authentication failed. Please check your API key in the .env file.")
+                    return
+                elif response.status_code == 403:
+                    print(f"Authorization failed. Your API key may not have access to this resource.")
+                    return
                 print(f"Error details: {response.text}")
                 # Retry on server errors (5xx) but not on client errors (4xx)
                 if response.status_code >= 500: